University of KentuckyUKnowledge

Journal Articles Center Endowed Faculty and DirectorPublications

1-1-1999

Physical Violence During Pregnancy: MaternalComplications and Birth OutcomesVilma E. CokkinidesAmerican Cancer Society

Ann L. CokerUniversity of Kentucky, ann.coker@uky.edu

Maureen SandersonMeharry Medical College, msanderson@mmc.edu

Cheryl AddyUniversity of South Carolina, caddy@sc.edu

Lesa Bethea

This Article is brought to you for free and open access by the Center Endowed Faculty and Director Publications at UKnowledge. It has been acceptedfor inclusion in Journal Articles by an authorized administrator of UKnowledge. For more information, please contact UKnowledge@lsv.uky.edu.

Repository CitationCokkinides, Vilma E.; Coker, Ann L.; Sanderson, Maureen; Addy, Cheryl; and Bethea, Lesa, "Physical Violence During Pregnancy:Maternal Complications and Birth Outcomes" (1999). Journal Articles. Paper 133.http://uknowledge.uky.edu/crvaw_facpub/133

Physical Violence During Pregnancy: MaternalComplications and Birth Outcomes

VILMA E. COKKINIDES, PhD, ANN L. COKER, PhD, MAUREEN SANDERSON, PhD,

CHERYL ADDY, PhD, AND LESA BETHEA, MD

Objective: To assess the association between physical vio-lence during the 12 months before delivery and maternalcomplications and birth outcomes.

Methods: We used population-based data from 6143women who delivered live-born infants between 1993 and1995 in South Carolina. Data on women’s physical violenceduring pregnancy were based on self-reports of “partner-inflicted physical hurt and being involved in a physicalfight.” Outcome data included maternal antenatal hospital-izations, labor and delivery complications, low birthweights, and preterm births. Odds ratios and 95% confidenceintervals were calculated to measure the associations be-tween physical violence, maternal morbidity, and birthoutcomes.

Results: The prevalence of physical violence was 11.1%.Among women who experienced physical violence, 54%reported having been involved in physical fights only and46% had been hurt by husbands or partners. In the lattergroup, 70% also reported having been involved in fighting.Compared with those not reporting physical violence,women who did were more likely to deliver by cesarean andbe hospitalized before delivery for maternal complicationssuch as kidney infection, premature labor, and trauma due tofalls or blows to the abdomen.

Conclusion: Physical violence during the 12 months beforedelivery is common and is associated with adverse maternalconditions. The findings support the need for research onhow to screen for physical violence early in pregnancy andto prevent its consequences. (Obstet Gynecol 1999;93:661–6.© 1999 by The American College of Obstetricians andGynecologists.)

Intimate partner violence against pregnant women isnot unusual. Studies have shown a prevalence of phys-

ical violence during pregnancy of between 0.9% and20%.1 Violence during pregnancy affects women andmight lead to pregnancy complications or adverse birthoutcomes.2,3

Violence can affect pregnancy through direct or indi-rect mechanisms.2,3 A blow to a pregnant woman’sabdomen can cause adverse outcomes directly (ie, fetalinjury and death, or complications such as pretermlabor). The indirect mechanism relates to a woman’svictimization experience from intimate-partner violenceand how it can induce intermediary risks (ie, psycho-logic stress or insufficient access to medical care) thatcould cause poor outcomes.2,3

Five studies4–8 showed no association between phys-ical violence and low birth weights (LBWs) or pretermbirths, whereas two others9,10 found modest increases inrisk of LBWs among women who experienced intimate-partner violence during pregnancy. Differing studypopulations, sample sizes, study designs, and varyingmeasures of physical violence might explain discrepantfindings among studies.

An understanding of the relationship between phys-ical violence during pregnancy, adverse maternal con-ditions, and birth outcomes could have important clin-ical and public health implications. Early identificationand intervention to prevent violence against pregnantwomen might reduce adverse outcomes in pregnancy.Using a population-based, cross-sectional study sam-ple, we assessed the independent association of physi-cal violence during pregnancy with maternal antenatalhospitalizations, labor and delivery complications,LBWs, and preterm births, while controlling for poten-tial confounders.

Materials and Methods

We analyzed data from the South Carolina PregnancyRisk Assessment Monitoring System,11 a population-based survey developed to monitor selected self-

From the Institute for Families in Society, Department of Epidemiol-ogy and Biostatistics, School of Public Health, and the Department ofFamily and Preventive Medicine, School of Medicine, University ofSouth Carolina, Columbia, South Carolina.

This research was funded in part by Collaborative AgreementU50/CC407132 from the Division of Reproductive Health, Centers forDisease Control and Prevention, Atlanta, Georgia.

661VOL. 93, NO. 5, PART 1, MAY 1999 0029-7844/99/$20.00PII S0029-7844(98)00486-4

Hosted in the Center for Research on Violence Against Women institutional repository with written permission from Wolters Kluwer Health.

reported maternal behaviors and experiences occurringbefore, during, and after pregnancy among women whodelivered live-born infants. A systematic birth weight–stratified random sample of new mothers (200–240women) was drawn monthly from the South Carolinabirth registry. Because of the low frequency of births ofLBW infants in the population, oversampling of LBWinfants is used to achieve large cell-sample size, relativeto normal birth weight infants. Between 1993 and 1995,9563 women 2 and 6 months postpartum were surveyedand 6718 participated (70% response). All mothers inthe sample were mailed a 14-page questionnaire (up tothree times) and monetary incentives were used toenhance response rates. If there was still no response,telephone follow-up was done (98% of women re-sponded by mail). Information in the analyses camefrom mothers’ completed surveys and their infants’birth certificates. We excluded 346 women with multi-ple gestations because their infants were at greater riskfor prematurity and LBW, 74 women who did notidentify themselves as black or white, 115 women withmissing data on physical violence, and 52 women withimplausible birth weight–gestational age combinations.The final sample consisted of 6143 women who hadsingleton live births between 1993 and 1995. The datawas weighted to adjust for nonresponse differentials bymaternal sociodemographic status and for samplingprobabilities. Weighted estimates were adjusted fornonresponse on the assumption that women in a par-ticular subgroup who responded represent those whodid not respond.

We defined physical violence during pregnancy us-ing two questions within an 18-item inventory of stress-ful life events.12 Women who reported being “physical-ly hurt by their husbands or partners” or “involved in aphysical fight” during the 12 months before deliverywere classified as having been exposed to physicalviolence during pregnancy. We reported elsewhere thatwomen who experienced partner physical violencewere similar to those who were involved in physicalfights during pregnancy.13 Women denying violentstressors constituted the comparison group. The surveydid not include additional questions on timing, sever-ity, and frequency of violence during pregnancy or anyquestions about emotional or sexual violence by part-ners during pregnancy.

In the survey, women were asked about hospitaliza-tions before delivery. We categorized their responses asdiabetes, high blood pressure, kidney infections, pre-mature labor, nausea, vomiting or dehydration, andvaginal bleeding or placenta problems. Specific condi-tions originally listed as “other” were recoded as otherinfections, loss of the infant, or trauma due to falls orblows to the abdominal region. Women who reported

no antenatal hospitalizations made up the referentgroup. For about 1.6% of the study sample, data onhospitalizations were missing. Birth certificates pro-vided information on complications of labor and deliv-ery (including febrile conditions, prolonged or dysfunc-tional labors, abruptio placentae, premature rupture ofmembranes [PROM], other excessive bleeding, and “fe-tal distress”) or cesarean delivery. Data on complica-tions of labor and delivery were missing for only fivewomen. From birth certificate data, birth weights wereclassified as less than 2500 g (LBW) and 2500 g or more.Infants whose gestational age was less than 37 weekswere considered preterm and those whose gestationalage was 37 weeks or more were considered term. Wealso combined birth weight and prematurity categoriesto create LBW-preterm, LBW-term, normal birthweight–preterm, and normal birth weight–term (com-parison group) categories.

Potential confounders of the association betweenviolence during pregnancy, maternal complications,and adverse birth outcomes were maternal race, mater-nal age, maternal education, poverty status, maritalstatus plus infant’s paternity information, number ofprevious children, adequacy of prenatal care utilizationindex,14 smoking and alcohol use during the last trimes-ter of pregnancy, and pregnancy intent at conception.

Simple categorical analyses and logistic regressionmodels using weighted survey data were conductedwith Survey Data Analysis (SUDAAN 7.4; ResearchTriangle Institute, Research Triangle Park, NC) soft-ware to adjust standard errors to account for selectionand response probabilities of the survey design. Weinitially examined the prevalence distribution of phys-ical violence during pregnancy. We compared the prev-alence of reporting physical violence with selectedmaternal characteristics and calculated the prevalenceodds ratios (ORs) and 95% confidence intervals (CIs)adjusted for potential confounders.

We used logistic regression to calculate ORs and 95%CIs to estimate the association between physical vio-lence during pregnancy (independent variable), mater-nal conditions, and birth outcomes. The dependentvariables in each of the respective models were mater-nal morbidity (antenatal maternal cause-specific hospi-talizations and specific complications of labor and de-livery) and birth outcomes (LBW and prematurity).These associations were adjusted for important con-founders, including age, poverty status, obtainment ofprenatal care, smoking during the third trimester, ma-ternal race, and parity. Confounders were selected forinclusion in the multivariate model if they were associ-ated with the respective outcome and physical violenceand if their addition to the model changed the crude ORby 10%.15

662 Cokkinides et al Violence During Pregnancy Obstetrics & Gynecology

Hosted in the Center for Research on Violence Against Women institutional repository with written permission from Wolters Kluwer Health.

Results

Eleven percent of women reported physical violence(either “being physically hurt by husband or partner”or “being involved in a physical fight”) within the 12months of delivering a live infant (Table 1). The prev-alence of being involved in a physical fight during theyear before delivery (9.6%) was twice as high as theprevalence of partner-perpetrated physical violence(5.1%). Also shown in Table 1 are the mutually exclu-sive categories of physical violence during pregnancy.Those data showed that more than half (54%) of thephysical violence was the result of being involved infights only, and 46% were hurt by husbands or partners.Of the women who reported partner-perpetrated phys-ical violence, 70% said they had been involved in fights.

Table 2 shows the weighted prevalence of reportingphysical violence by select sociodemographic and ma-ternal behavioral characteristics, along with the ad-justed prevalence ORs between physical violence andselect maternal characteristics. Because some of thosefindings were described elsewhere,13 we will summa-rize the results briefly. Women with greater odds ofphysical violence have characteristics associated withgreater risks for adverse pregnancy outcomes (eg,young age, poverty, unmarried status, unwanted preg-nancy, and smoking and drinking alcohol during preg-nancy).

The prevalence distribution of self-reported maternalmorbidity among women who experienced violenceduring pregnancy is presented in Table 3, along withthe adjusted ORs with 95% CIs for the associationbetween violence and those maternal conditions ad-justed for confounding. Among women who had expe-

Table 1. Prevalence of Physical Violence During the 12Months Before Delivery

Characterization of physicalviolence during pregnancy

Unweighted no.(n 5 6143) Prevalence*

Physically hurt by husbandor partner

330 5.1 (2.0, 7.1)

Involved in physical fight 583 9.6 (8.2, 10.2)Involved in physical fight

or physically hurt byhusband or partner

680 11.1 (10.3, 12.4)

Involved in a physical fightonly; not physically hurtby husband or partner

350 6.0 (5.2, 7.0)

Physically hurt by husbandor partner only; notinvolved in physical fight

97 1.5 (1.1, 1.9)

Physically hurt by husbandor partner and involvedin physical fight

233 3.6 (2.9, 4.3)

* Weighted prevalence percent and 95% confidence interval ofprevalence percent.

Table 2. Prevalence of Physical Violence During Pregnancy,by Selected Maternal Characteristics

Correlate

Physicalviolence

(n 5 680)

Nophysicalviolence

(n 5 5463)OR*

(95% CI)

RaceBlack 16.2% (1.1) 83.8% (1.1) 1.3 (0.9, 1.7)White 7.9% (0.6) 92.1% (0.6) 1.0

Age (y),20 21.8% (1.8) 78.2% (1.8) 3.0 (1.9, 4.8)20–29 11.0% (0.8) 89.0% (0.8) 1.8 (1.2, 2.7).29 4.6% (0.7) 95.4% (0.7) 1.0

Education (y),12 20.8% (1.6) 79.2% (1.6) 1.8 (1.2, 2.8)12 11.7% (0.9) 88.3% (0.9) 1.3 (0.8, 1.8)$13 5.4% (0.7) 94.6% (0.7) 1.0

Poverty (poverty rate)#100% 17.4% (1.0) 82.6% (1.0) 5.6 (3.4, 9.1).100–185% 9.9% (1.3) 90.1% (1.3) 3.2 (1.9, 5.5).185% 2.8% (0.5) 97.2% (0.5) 1.0

Marital status plus infant’spaternity informationUnmarried with

paternity data missing20.1% (1.3%) 79.9% (1.3) 2.6 (1.8, 3.8)

Married with paternitydata refused

37.4% (6.3) 62.6% (6.3) 8.4 (4.2, 16.9)

Unmarried withpaternityacknowledgment

18.0% (2.6) 82.0% (2.6) 2.5 (1.5, 4.1)

Married with paternitydata

4.8% (0.5) 95.2% (0.5) 1.0

Number of previouschildren$3 17.2% (2.9) 82.8% (2.9) 2.3 (1.3, 3.9)2 11.2% (1.5) 88.8% (1.5) 1.4 (0.9, 2.2)1 10.7% (1.0) 89.3% (1.0) 1.3 (1.0, 1.8)0 10.9% (0.8) 89.1% (0.8) 1.0

Adequacy of obtainmentof prenatal careInadequate/no prenatal

care19.8% (2.0) 80.2% (2.0) 1.3 (0.9, 1.9)

Intermediate 8.0% (1.3) 92.0% (1.3) 0.8 (0.5, 1.2)Adequate plus

(intensive)11.0% (1.1) 89.0% (1.1) 1.2 (0.8, 1.6)

Adequate 9.9% (0.7) 90.1% (0.7) 1.0Smoking/alcohol-drinking

status during 3rdtrimesterSmoked and drank 21.9% (5.9) 78.1% (5.9) 3.7 (1.6, 8.4)Smoked or drank 17.5% (1.7) 82.5% (1.7) 2.0 (1.5, 2.8)Neither smoked nor

drank9.8% (0.6) 90.2% (0.6) 1.0

Pregnancy intentUnwanted 21.3% (2.1) 78.7% (2.1) 2.3 (1.6, 3.4)Mistimed 14.0% (1.1) 86.0% (1.1) 1.4 (1.0, 1.9)Wanted 6.6% (0.6) 93.4% (0.6) 1.0

OR 5 odds ratio; CI 5 confidence interval.Data are presented as % (standard error).Percentages are weighted to account for sampling scheme.* Odds ratios were adjusted for age and poverty.

VOL. 93, NO. 5, PART 1, MAY 1999 Cokkinides et al Violence During Pregnancy 663

Hosted in the Center for Research on Violence Against Women institutional repository with written permission from Wolters Kluwer Health.

rienced violence, 27.2% reported antenatal hospitaliza-tions before delivery. After adjustment for confoundingfactors, the following maternal conditions leading tohospitalizations were found to be associated with re-porting physical violence: kidney infections (OR 2.7,95% CI 1.3, 5.5), premature labor (OR 1.8, 95% CI 1.2,2.7), and trauma due to falls or blows to the abdomen(OR 20.2, 95% CI 1.9, 206.0).

The prevalence of complications of labor and deliveryand birth outcomes are given in Table 4. After control-ling for age, poverty, obtainment of prenatal care, andsmoking status during pregnancy, we found thatwomen who experienced physical violence duringpregnancy were 1.5 times more likely to deliver bycesarean. Physical violence was not significantly asso-ciated with risk of LBW, preterm birth, or LBW–premature birth combinations.

Discussion

Violence against women is difficult to measure forvarious reasons, including the fact that a standardizeddefinition is lacking, and the fact that some women areunwilling to disclose violence, because of social stigmaor cultural sanctioning of violence.16 Thus, the bestprevalence estimates are probably underestimates.Given this and the similar effects on maternal outcomesof being involved in a fight, we included data onwomen’s involvement in fights in our assessment ofphysical violence against pregnant women. Fightingoften leads to injuries, which range from minor to

serious. Our estimate of physical violence during preg-nancy (11%) falls within the range of prevalence esti-mates reported elsewhere.1

Of the self-reported maternal conditions leading toantenatal hospitalizations, kidney infections, prematurelabor, and trauma due to falls or blows to the abdomi-nal region were associated with physical violence dur-ing pregnancy. We believe that a direct mechanism ismost likely indicated by those observations. We arelimited in trying to explicate further this mechanism ofaction because we do not know the timing or severity ofviolence. Physical violence during pregnancy is thesecond leading cause of trauma during pregnancy, aftermotor vehicle accidents.17 In our study, too few womenreported trauma to the abdominal area (as expected,given that our study sample was population based andnot taken from an emergency department population),so we lacked power to address adequately the associa-tion between physical violence and that adverse out-come. Trauma research studies showed that severephysical trauma to the maternal abdomen might lead tohospitalizations that might cause premature labor ordelivery.10,18,19 Many women in abusive intimate rela-tionships also experience sexual assault4,10,16 and are atincreased risk of contracting sexually transmitted dis-eases (STDs) from their partners.4 Research establishedan association between STDs and preterm labor.20 Wefound that reporting antenatal hospitalization for pre-mature labor was significantly associated with physicalviolence during pregnancy. That finding might be im-portant for clinical management of high-risk patients by

Table 3. Association Between Maternal Morbidity and Physical Violence During Pregnancy

Maternal hospitalization

Physical violence(n 5 680)

No physical violence(n 5 5463)

OR* (95% CI)No. % No. %

Any hospitalization before delivery 269 27.2% (2.4) 1661 18.3% (0.7) 1.5 (1.1, 2.0)Hospitalization for

Diabetes 6 1.3% (0.8) 59 0.9% (0.2) 1.3 (0.3, 5.7)High blood pressure 54 6.4% (1.5) 422 3.6% (0.3) 1.8 (1.0, 3.4)Kidney infection 32 6.4% (1.6) 112 1.9% (0.3) 2.7 (1.3, 5.5)Premature labor 145 14.2% (2.0) 772 7.5% (0.5) 1.8 (1.2, 2.7)Nausea, vomiting, or

dehydration33 5.8% (1.4) 170 3.2% (0.4) 1.8 (0.9, 3.7)

Vaginal bleeding, placentaproblems

45 2.0% (0.6) 269 2.2% (0.3) 0.8 (0.3, 1.9)

Other infections† 11 2.7% (1.0) 45 1.0% (0.2) 2.3 (0.9, 6.4)Miscarriage† 1 0.01% (0.01) 4 0.01% (0.01) 2.3 (0.5, 13.3)Trauma due to fall or blow to

stomach area†

4 1.2% (0.7) 8 0.1% (0.04) 20.2 (1.9, 206.0)

Abbreviations as in Table 2.Data are presented as n or % (standard error).Comparison group was women who were not hospitalized before delivery.* Adjusted for age, poverty, obtainment of prenatal care, and smoking during pregnancy.† Categories were developed through recoding of conditions originally listed as “other.”

664 Cokkinides et al Violence During Pregnancy Obstetrics & Gynecology

Hosted in the Center for Research on Violence Against Women institutional repository with written permission from Wolters Kluwer Health.

obstetricians and perinatologists because preterm laborcorrelates with premature birth. Preterm birth and LBWare among the leading causes of perinatal morbidityand mortality.21 We also showed a significant associa-tion between physical violence and kidney infections.The effect of physical violence on pregnant womenappears to be extensive and such violence is likely toaffect every organ system.2

This report showed that physical violence is associ-ated with maternal smoking and drinking of alcoholduring pregnancy, which was corroborated by otherstudies.2,3,4,13 Such risky behavior might be connectedto greater levels of stress associated with violence-related victimization,2,3,16 which suggests the need forearly identification of intimate-partner violence amongpregnant women in substance abuse treatment andcessation programs and the need for appropriate inter-ventions between those treatment programs and shel-ters for persons who have experienced domestic vio-lence.

We found an association between physical violenceduring pregnancy and cesarean delivery, independentof other confounding factors (age, poverty, obtainmentof prenatal care, and smoking during pregnancy). Wecannot rule out other potential confounders, such asprevious cesarean delivery and malpresentation, in thisassociation. However, the observed relationship might

be due to the fact that women reporting physicalviolence have various maternal complications and un-dergo hospitalizations before delivery and thus aremonitored medically for untoward outcomes.

After adjustments for maternal age, poverty, involve-ment in prenatal care, and maternal smoking duringpregnancy, we found no association between physicalviolence during pregnancy and LBW or prematurity.Although those findings are consistent with those ofother studies,4–8 two studies9,10 found positive associa-tions between physical violence and LBW. More re-search is needed to determine how violence can affectbirth outcomes.

Our study strengths were the use of a population-based sample that allowed inferences to the entirepopulation of women who had live-born infants, highresponse rates, and many data on women that served aspotential confounders. However, several factors limitedthe interpretation of our findings. The cross-sectionaldesign precluded examining physical violence as anantecedent factor to some outcomes. Some womenmight have underreported their physical violence expe-riences during pregnancy. We did not assess physicalviolence by timing, severity, and frequency. The self-reported maternal hospitalizations and data on laborand delivery complications from birth certificates couldnot be verified using medical records. Other important

Table 4. Association Between Maternal Complications During Labor, Delivery and Birth Outcomes, and Physical ViolenceDuring Pregnancy

Outcome

Physical violence(n 5 680)

No physical violence(n 5 5463)

OR(95% CI)No. % No. %

Complication of labor and deliveryHad complication of labor and

delivery*323 37.7% (2.6) 2507 36.7% (0.9) 1.1† (0.8, 1.4)

Cesarean delivery* 198 25.5% (2.4) 1575 21.8% (0.8) 1.5† (1.1, 2.0)Premature rupture of

membranes*72 2.7% (1.0) 496 4.3% (0.4) 2.0† (0.7, 7.3)

Fetal distress* 44 8.3% (1.6) 272 5.3% (0.6) 1.8‡ (1.0, 3.3)Birth outcome

Low birth weight§ 327 9.4% (0.8) 2388 7.8% (0.1) 1.0\ (0.8, 1.2)Preterm¶ 275 11.1% (1.4) 1954 8.2% (0.4) 1.2%\ (0.9, 1.7)

Combination of low birth weight–prematurity

Preterm–low birth weight** 256 5.8% (0.5) 1846 4.9% (0.1) 1.0\ (0.7, 1.3)Preterm–normal birth weight 19 5.3% (1.3) 108 3.3% (0.3) 1.9 (1.0, 3.7)Term–low birth weight 71 3.6% (0.5) 513 2.9% (0.1) 1.0 (0.7, 1.4)

Abbreviations as in Table 2.Data are presented as n or % (standard error).* Comparison group was women who did not have any complications of labor and delivery.† Adjusted for age, poverty, obtainment of prenatal care, and smoking during pregnancy.‡ Adjusted for age, poverty, and obtainment of prenatal care.§ Comparison group was women with normal birth weight infants.\ Adjusted for age, poverty, obtainment of prenatal care, and race.¶ Comparison group was women with term infants.** Comparison group was women with normal birth weight–term infants.

VOL. 93, NO. 5, PART 1, MAY 1999 Cokkinides et al Violence During Pregnancy 665

Hosted in the Center for Research on Violence Against Women institutional repository with written permission from Wolters Kluwer Health.

confounders were not assessed, including illicit druguse. Those factors might lead to nondifferential misclas-sification, which would result in underestimation of theassociation of physical violence during pregnancy andthose outcomes. In future studies, attempts should bemade specifically to assess not only physical violencebefore and during pregnancy but also sexual and emo-tional violence and the potential effects of these forms ofviolence on maternal and birth outcomes.

Health professionals must regard violence duringpregnancy as a serious public health problem, onerequiring the same alertness currently focused on ges-tational diabetes, preeclampsia, and PROM. Routinescreening has been recommended at the various entrypoints of contact between pregnant women and medicalcare (eg, primary care, prenatal care, obstetric, andgynecologic services).22

References1. Gazmararian JA, Lazorick S, Spitz AM, Ballard TJ, Saltzman LE,

Marks JS. Prevalence of violence against pregnant women. JAMA1996;275:1915–20.

2. Newberger EH, Barkan SE, Lieberman ES, McCormick MC, Yllo K,Gary LT, et al. Abuse of pregnant women and adverse birthoutcome. Current knowledge and implications for practice. JAMA1992;267:2370–2.

3. Petersen R, Gazmararian J, Spitz A, Rowley DL, Goodwin MM,Saltzman LE, et al. Violence and adverse pregnancy outcomes: Areview of the literature and directions for future research. Am JPrev Med 1997;13:366–73.

4. Amaro H, Fried LE, Cabral H, Zuckerman B. Violence duringpregnancy and substance use. Am J Public Health 1990;80:575–9.

5. Berenson AB, Wiemann CM, Wilkinson GS, Jones WA, AndersonGD. Perinatal morbidity associated with violence experienced bypregnant women. Am J Obstet Gynecol 1994;170:1760–9.

6. Grimstad H, Schei B, Backe B, Jacobsen G. Physical abuse and lowbirthweight: A case-control study. Br J Obstet Gynaecol 1997;104:1281–7.

7. O’Campo P, Gielen AC, Faden RR, Kass N. Verbal abuse andphysical violence among a cohort of low-income pregnant women.Womens Health Issues 1994;4:29–37.

8. Schei B, Samuelsen SO, Bakketeig LS. Does spousal physical abuseaffect the outcome of pregnancy? Scand J Soc Med 1991;19:26–31.

9. Bullock LF, McFarlane J. The birth-weight/battering connection.Am J Nurs 1989;89:1153–5.

10. Parker B, McFarlane J, Soeken K. Abuse during pregnancy: Effectson maternal complications and birth weight in adult and teenagewomen. Obstet Gynecol 1994;84:323–8.

11. Adams MM, Shulman HB, Bruce C, Hogue C, Brogan D. ThePregnancy Risk Assessment Monitoring System: Design, question-naire, data collection and response rates. PRAMS Working Group.Paediatr Perinat Epidemiol 1991;5:333–46.

12. Newton RW, Hunt LP. Psychosocial stress in pregnancy and itsrelation to low birthweight. BMJ 1984;288:1191–4.

13. Cokkinides VE, Coker AL. Experiencing physical violence duringpregnancy: Prevalence and correlates. Fam Community Health1998;20:19–37.

14. Alexander GR, Kotelchuck M. Quantifying the adequacy of prena-tal care: A comparison of indices. Public Health Rep 1996;111:408–18.

15. Breslow N, Day N. The design and analysis of case-control studies.Statistical methods in cancer research. IARC Scientific Publicationsno. 32 (vol 1). Oxford, UK: Oxford University Press, 1984.

16. Stark E, Flitcraft A. Women at risk: Domestic violence and wom-en’s health. Thousand Oaks, California: Sage, 1996.

17. Connolly AM, Katz VL, Bash KL, McMahon MJ, Hansen WF.Trauma and pregnancy. Am J Perinatol 1997;14:331–6.

18. Williams MA, Lieberman E, Mittendorf R, Monson RR, Schoen-baum SC. Risk factors for abruptio placentae. Am J Epidemiol1991;134:965–72.

19. Webster J, Chandler J, Battistitta D. Pregnancy outcomes andhealth care use: Effects of abuse. Am J Obstet Gynecol 1996;174:760–7.

20. Reynolds HD. Bacterial vaginosis and its implications in pretermlabor and premature rupture of membranes. A review of theliterature. J Nurse Midwifery 1991;36:289–96.

21. Institute of Medicine. Preventing low birthweight. Washington,DC: National Academic Press, 1985.

22. Campbell JC. Abuse during pregnancy: Progress, policy, andpotential. Am J Public Health 1998;88:185–7.

Address reprint requests to:Vilma E. Cokkinides, PhDInstitute for Families in SocietyUniversity of South Carolina937 Assembly Street, Carolina Plaza 12th FloorColumbia, SC 29208E-mail: vilmac@gwm.sc.edu

Received July 16, 1998.Received in revised form September 28, 1998.Accepted October 15, 1998.

Copyright © 1999 by The American College of Obstetricians andGynecologists. Published by Elsevier Science Inc.

666 Cokkinides et al Violence During Pregnancy Obstetrics & Gynecology

Hosted in the Center for Research on Violence Against Women institutional repository with written permission from Wolters Kluwer Health.